Reproduce “StableGrasp: Reconstructing Physically Stable Human Hand Grasps from Single Images” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Han Jiang, Etienne Vouga, Qixing Huang, Georgios Pavlakos
StableGrasp reconstructs monocular hand-object grasps by optimizing both an image-aligned hand pose and a separate target for a PD controller. It minimizes simulated hand-object kinetic energy while regularizing pose and contact. In forward physics simulation on DexYCB, it achieves 87.45% success under a 1 cm motion threshold, versus 64.02% for Pinching; this does not establish real-world performance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.